Serve the marketplace to Codex and ChatGPT as well as Claude Code

.agents/plugins/marketplace.json

@@ -0,0 +1,22 @@
+{
+  "name": "audioshake",
+  "interface": {
+    "displayName": "AudioShake"
+  },
+  "plugins": [
+    {
+      "name": "audioshake-cli",
+      "source": {
+        "source": "git-subdir",
+        "url": "https://github.com/audioshake/audioshake-python.git",
+        "path": "plugin",
+        "ref": "main"
+      },
+      "policy": {
+        "installation": "AVAILABLE",
+        "authentication": "ON_USE"
+      },
+      "category": "Creativity"
+    }
+  ]
+}

.claude-plugin/marketplace.json

@@ -1,20 +1,20 @@
 {
   "name": "audioshake",
-  "description": "AudioShake plugins and skills for Claude: stem separation, transcription and more.",
+  "description": "AudioShake plugins and skills for AI coding agents: stem separation, transcription and more.",
   "owner": {
     "name": "AudioShake",
     "url": "https://www.audioshake.ai"
   },
   "plugins": [
     {
-      "name": "audioshake",
-      "displayName": "AudioShake",
-      "description": "Run AudioShake models from Claude with the audioshake CLI and Python library: stems, dialogue isolation, speech clean-up, lyric transcription and alignment.",
+      "name": "audioshake-cli",
+      "displayName": "AudioShake CLI",
+      "description": "Run AudioShake models with the audioshake CLI and Python library: stems, dialogue isolation, speech clean-up, lyric transcription and alignment.",
       "source": {
         "source": "git-subdir",
         "url": "audioshake/audioshake-python",
         "path": "plugin",
-        "ref": "v0.1.0"
+        "ref": "main"
       },
       "author": { "name": "AudioShake" },
       "homepage": "https://github.com/audioshake/audioshake-python",

README.md

@@ -1,34 +1,49 @@
-# AudioShake for Claude
+# AudioShake plugins
 
-AudioShake plugins and skills for [Claude Code](https://claude.com/claude-code).
-Ask Claude for stems, an instrumental, clean dialogue or lyrics with word timings,
+AudioShake plugins and skills for AI coding agents: [Claude Code](https://claude.com/claude-code),
+[Codex](https://developers.openai.com/codex) and the ChatGPT desktop app.
+Ask your agent for stems, an instrumental, clean dialogue or lyrics with word timings,
 and it runs [AudioShake](https://www.audioshake.ai) models for you.
 
 ## Install
 
+You need an AudioShake account. Create an API key in the
+[AudioShake dashboard](https://dashboard.audioshake.ai) and export it as
+`AUDIOSHAKE_API_KEY`, or let your agent walk you through `audioshake login`.
+
+### Claude Code
+
 Inside Claude Code:
 
 ```
 /plugin marketplace add audioshake/claude-plugins
-/plugin install audioshake@audioshake
+/plugin install audioshake-cli@audioshake
 ```
 
 Or from your shell:
 
 ```bash
 claude plugin marketplace add audioshake/claude-plugins
-claude plugin install audioshake@audioshake
+claude plugin install audioshake-cli@audioshake
 ```
 
-You need an AudioShake account. Create an API key in the
-[AudioShake dashboard](https://dashboard.audioshake.ai) and export it as
-`AUDIOSHAKE_API_KEY`, or let Claude walk you through `audioshake login`.
+### Codex
+
+From your shell:
+
+```bash
+codex plugin marketplace add audioshake/claude-plugins
+codex plugin add audioshake-cli@audioshake
+```
+
+Inside a Codex session, `/plugins` opens the same catalog. In the ChatGPT
+desktop app, open the Plugins Directory and choose the AudioShake marketplace.
 
 ## Plugins
 
 | Plugin | What it does | Source |
 | --- | --- | --- |
-| `audioshake` | Runs AudioShake models through the `audioshake` CLI and Python library: stem separation, dialogue and effects isolation, speech clean-up, multi-speaker separation, music detection and removal, lyric transcription and alignment. | [audioshake-python](https://github.com/audioshake/audioshake-python/tree/main/plugin) |
+| `audioshake-cli` | Runs AudioShake models through the `audioshake` CLI and Python library: stem separation, dialogue and effects isolation, speech clean-up, multi-speaker separation, music detection and removal, lyric transcription and alignment. | [audioshake-python](https://github.com/audioshake/audioshake-python/tree/main/plugin) |
 
 Once installed, just ask:
 
@@ -37,18 +52,33 @@ Once installed, just ask:
 - "Pull the dialogue out of `scene.mp4` and drop the music."
 - "Transcribe the lyrics of `track.mp3` with word timings."
 
-Every run costs credits, charged per minute of audio for each model. Claude
+Every run costs credits, charged per minute of audio for each model. Your agent
 checks with you before large batches.
 
 ## Updates
 
-Each plugin is pinned to a tested release of the project it comes from. To get
-the latest:
+Each plugin entry points at a release or branch of the project it comes from.
+To get the latest:
 
 ```
-/plugin marketplace update audioshake
+/plugin marketplace update audioshake      # Claude Code
+codex plugin marketplace upgrade audioshake  # Codex
 ```
 
+## How this repository is laid out
+
+One catalog, two manifests. Both list the same plugins from the same sources.
+
+| File | Read by |
+| --- | --- |
+| `.claude-plugin/marketplace.json` | Claude Code |
+| `.agents/plugins/marketplace.json` | Codex CLI, Codex IDE extensions, ChatGPT desktop |
+
+Each plugin's own directory carries a `.claude-plugin/plugin.json` and a
+`.codex-plugin/plugin.json`, with a shared `skills/` folder in the open
+[Agent Skills](https://agentskills.io) format. The skills also install directly
+into any agent that reads that format, with no marketplace involved.
+
 ## Issues
 
 Report a problem with a plugin in its source repository, linked in the table